Patent Document 1 discloses a three-dimensional image constructing method for perspectively coordinate-transforming an examinee having an internal space to a coordinate system on any point-of-view plane to obtain a two-dimensional image in a volume image obtained by stacking many tomographic images, and shading the two-dimensional image while the distance between a virtual line light source provided in the examinee and a contour point of the inside of the examinee and the distance between the outside of the examinee and the point-of-view plane are reflected to the two-dimensional image, thereby obtaining a three-dimensional image in which shading corresponding to the inner surface shape and the outer surface shape of the examinee is applied to the outer surface of the examinee.